Target Identity Probability Analysis for Multi-Target Tracking

Shaoming He*, Hyo Sang Shin, Antonios Tsourdos

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes a novel target identity probability analysis algorithm for multi-target tracking (MTT) problems. By introducing a target identity probability matrix, finding the identity probability is converted into the problem of target-to-target association. The issue in solving the equivalent problem is that the computational complexity increases exponentially with the increase of the problem size. We propose a Gibbs sampling approach to find approximate polynomial-time solutions. Theoretical analysis reveals that the proposed algorithm provides a performance-guaranteed approximation. Numerical simulations are conducted to support the analytical findings.

Original languageEnglish
Title of host publicationProceedings of the 35th Chinese Control and Decision Conference, CCDC 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2459-2465
Number of pages7
ISBN (Electronic)9798350334722
DOIs
Publication statusPublished - 2023
Event35th Chinese Control and Decision Conference, CCDC 2023 - Yichang, China
Duration: 20 May 202322 May 2023

Publication series

NameProceedings of the 35th Chinese Control and Decision Conference, CCDC 2023

Conference

Conference35th Chinese Control and Decision Conference, CCDC 2023
Country/TerritoryChina
CityYichang
Period20/05/2322/05/23

Keywords

  • Information-Driven Joint Probabilistic Data Association
  • Multi-Target Tracking
  • Target Identity Probability

Fingerprint

Dive into the research topics of 'Target Identity Probability Analysis for Multi-Target Tracking'. Together they form a unique fingerprint.

Cite this